JP7393260B2 - estimation device - Google Patents

Description

The present invention relates to an estimation device for estimating the degree of risk of contact between a moving object and a traffic participant in front of the moving object.

Patent Document 1 discloses a vehicle control device that starts and accelerates the vehicle to avoid contact with other traffic participants (oncoming vehicles, pedestrians, etc.) when the vehicle changes course and passes through an intersection. It will be done. Specifically, this vehicle control device calculates the position, moving direction, and speed of the traffic participant, identifies the intersection position of the predicted course of the own vehicle and the predicted course of the traffic participant, and determines the intersection between the own vehicle and the traffic participant. Start and accelerate your vehicle so that the driver and the driver do not pass through the intersection at the same time.

JP 2017-121933 Publication

Like the device of Patent Document 1, the process of specifying the intersection position of the predicted course of the own vehicle and the predicted course of the traffic participant requires a large amount of calculation. Therefore, as the number of traffic participants increases, the computational load increases.

The present invention has been made in consideration of such problems, and an object of the present invention is to provide an estimation device that can reduce the calculation load.

The first aspect of the present invention is
an external world recognition unit that recognizes the surrounding environment of the moving object;
a distance acquisition unit that acquires a first distance from the moving body and a second distance longer than the first distance;
A scene defined by a road configuration, a position of the moving object, a position of a traffic participant around the moving object, and a degree of risk regarding contact between the moving object and the traffic participant are stored in association with each other. a scene memory section;
For the traffic participant whose separation distance between the mobile object and the traffic participant is less than or equal to the first distance, the speed of the traffic participant, the acceleration of the traffic participant, and the traveling direction of the traffic participant. A first estimation is performed to estimate the degree of risk based on the second distance, and for the traffic participants whose separation distance is greater than or equal to the second distance, the degree of risk is estimated based on the scene at that time. an estimation unit that performs a second estimation;
Equipped with

The second aspect of the present invention is
an external world recognition unit that recognizes the surrounding environment of the moving object;
a time calculation unit that calculates a scheduled approach time for the moving object and a traffic participant on the opposite lane to approach;
A scene defined by a road configuration, a position of the moving object, a position of the traffic participant around the moving object, and a degree of risk regarding contact between the moving object and the traffic participant are stored in association with each other. a scene memory section to
For the traffic participant whose scheduled approach time is less than or equal to a first time, the risk level is estimated based on the speed of the traffic participant, the acceleration of the traffic participant, and the traveling direction of the traffic participant. A first estimation is performed to estimate the degree of risk based on the scene at that time for the traffic participant whose estimated approach time is a second time or more, which is longer than the first time. an estimation unit that performs estimation;
Equipped with

According to the present invention, it is possible to reduce the load of calculations performed to estimate the degree of risk regarding contact between a moving object and a traffic participant.

FIG. 1 is a block diagram of a mobile object control device including an estimation device. FIG. 2 is a functional block diagram of the arithmetic device of the first embodiment. FIG. 3 is a diagram showing scene-risk information. 4A to 4E are diagrams showing specific examples of each scene in FIG. 3. FIG. 5 is a diagram showing a situation in which the degree of risk is estimated. FIG. 6 is a flowchart of the main processing. FIG. 7 is a flowchart of the execution determination process. FIG. 8 is a flowchart of the D1 correction process. FIG. 9 is a flowchart of the estimation process. FIG. 10 is a flowchart of execution determination processing that is different from FIG. 7. FIG. 11 is a flowchart of target lane determination processing. FIG. 12 is a functional block diagram of the arithmetic device of the second embodiment.

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an estimation device according to the present invention will be described in detail by citing preferred embodiments and referring to the accompanying drawings.

[1. First embodiment]
[1.1. Configuration of mobile object control device 10 and estimation device 36]
The configuration of the mobile object control device 10 including the estimation device 36 will be described using FIG. 1. The mobile body control device 10 is provided in a first mobile body 120 (FIG. 5). The first moving body 120 is, for example, a vehicle (own vehicle). The mobile object control device 10 has a so-called driving support function or automatic driving function that controls the speed V and steering of the first mobile object 120 regardless of the driver's intention.

The mobile object control device 10 operates the first mobile object 120 based on a main control device 12, a group of input devices that input various information to the main control device 12, and various information outputted by the main control device 12. and an output device group. The input device group includes an external world sensor 14, a navigation device 16, a positioning device 18, a receiving device 20, a vehicle behavior sensor 22, and an operation sensor 24. The output device group includes a drive device 28, a brake device 30, a steering device 32, and an HMI 34. The estimation device 36 according to the present embodiment includes, for example, the main control device 12, the external sensor 14, the navigation device 16, the positioning device 18, the receiving device 20, and the vehicle behavior sensor 22.

[1.1.1. Configuration of input device group]
The external sensor 14 includes multiple cameras 40, multiple radars 42, and multiple LiDARs 44. The camera 40 photographs the surroundings of the first moving body 120 and outputs image information to the main control device 12. The radar 42 and LiDAR 44 detect targets around the first moving body 120 and output detection information to the main controller 12.

The navigation device 16 uses GPS to measure the position of the first moving body 120 and generates a planned travel route from the position of the first moving body 120 to the destination specified by the driver. The navigation device 16 outputs route information indicating the generated planned travel route to the main control device 12.

The positioning device 18 includes a GNSS 46, an IMU 48, and a map DB 50. The positioning device 18 measures the position of the first mobile body 120 using the GNSS 46 and the IMU 48, and outputs position information indicating the position of the first mobile body 120 to the main controller 12. Furthermore, the positioning device 18 outputs map information stored in the map DB 50 to the main control device 12. Note that the map information stored in the map DB 50 is more accurate than the map information stored in the navigation device 16, and includes various information (information for each lane, etc.).

The receiving device 20 includes first to third receiving terminals (not shown). The first receiving terminal receives wide area information broadcast by a broadcasting station. The second receiving terminal receives local information transmitted by a roadside device installed on the road 130 (FIG. 5). The third receiving terminal receives information transmitted by traffic participants other than the first mobile object 120. The first to third receiving terminals output the received various information to the main control device 12.

The vehicle behavior sensor 22 includes sensors for measuring the behavior (velocity V, acceleration, yaw rate, etc.) of the first moving body 120. Each sensor outputs various detected information to the main controller 12.

The operation sensor 24 includes an automated switch 52 . The automation switch 52 outputs instruction information to the main controller 12 instructing automation or cancellation of automation of control of either the speed V or the steering, in response to a switch operation performed by the driver. Further, the operation sensor 24 includes various sensors that detect the amount of operation of driving operators (accelerator pedal, brake pedal, steering wheel).

[1.1.2. Configuration of main controller 12]
The main control device 12 is configured by an ECU. The main control device 12 includes an input/output device 56, an arithmetic device 58, and a storage device 60. The input/output device 56 includes an A/D conversion circuit, a communication interface, and the like. The arithmetic device 58 has a processor such as a CPU, for example. The arithmetic device 58 implements various functions by executing programs stored in the storage device 60. Various functions of the arithmetic unit 58 will be explained in [1.1.4] below. The storage device 60 includes a RAM, a ROM, and the like. The storage device 60 stores various programs and numerical information such as threshold values used in processing performed by the arithmetic device 58. The storage device 60 also stores scene-risk information 88. The scene-risk information 88 will be explained in [1.1.5] below.

[1.1.3. Configuration of output device group]
The drive device 28 includes a driving force output ECU and a control target of the driving force output ECU (both not shown). The drive device 28 adjusts the driving force according to the instruction information (drive instruction) output by the main control device 12.

The brake device 30 includes a brake ECU and a control target of the brake ECU (both not shown). The braking device 30 adjusts the braking force according to instruction information (braking instruction) output by the main control device 12.

The steering device 32 includes an EPS (electric power steering) ECU and an object controlled by the EPS ECU (both not shown). The steering device 32 adjusts the amount of steering according to instruction information (steering instruction) output by the main control device 12.

The HMI 34 includes a display device 62 and an audio device 64. The display device 62 outputs images in accordance with instruction information (notification instructions) output by the main control device 12. The audio device 64 outputs audio according to instruction information (notification instruction) output by the main control device 12.

[1.1.4. Various functions of arithmetic unit 58]
Various functions realized by the arithmetic device 58 will be explained using FIG. 2. The calculation device 58 functions as a control state setting section 66, a manual control section 68, an external world recognition section 70, a self-position recognition section 72, an action planning section 74, a mobile object control section 76, and an information control section 78. do.

The control state setting unit 66 determines whether various travel controls (speed V control and steering control) are to be performed manually or automatically, depending on the operation performed with the automation switch 52.

The manual control unit 68 performs travel control related to manual control according to the operation amount of the driving operation elements (accelerator pedal, brake pedal, steering wheel) output by the operation sensor 24. The manual control unit 68 outputs instruction information related to manual control (driving instructions, braking instructions, steering instructions) to the drive device 28, the braking device 30, and the steering device 32.

The external world recognition unit 70 recognizes the environment around the first moving body 120 based on the image information and detection information output by the external world sensor 14. The self-position recognition unit 72 recognizes the position of the first moving body 120 based on the position information and map information output by the positioning device 18.

The action planning section 74 creates an action plan related to automatic control based on the recognition result of the external world recognition section 70 and the recognition result of the self-position recognition section 72. For example, the action planning unit 74 generates a local map (dynamic map) that includes static information and dynamic information around the first moving object 120. Then, the action planning unit 74 determines the optimal action based on the local map and the state (velocity V, steering angle, position) of the first mobile object 120, and determines the speed V and travel trajectory for realizing the action. seek.

The action planning section 74 includes a distance acquisition section 84 and an estimation section 86. The distance acquisition unit 84 and the estimation unit 86 perform processing related to estimating the degree of risk. The distance acquisition unit 84 acquires a first distance D1 from the first moving object 120 and a second distance D2 that is longer than the first distance D1. The estimation unit 86 calculates the separation distance D between the first moving object 120 and a traffic participant in front of the first moving object 120, and estimates the degree of risk according to the separation distance D. The action planning section 74 determines whether the first moving object 120 should proceed or stop according to the degree of risk estimated by the estimating section 86, and determines an appropriate speed V when moving.

In this embodiment, the traffic participant in front of the first moving object 120 refers to the second moving object 122 that is traveling toward the intersection 132 into which the first moving object 120 is entering, as shown in FIG. means. Furthermore, in this embodiment, the degree of risk refers to the possibility that the first moving body 120 and the second moving body 122 will come into contact at the intersection 132. In this embodiment, the degree of risk is simply divided into two values. The risk level is set to "1" when there is a possibility that the first moving body 120 and the second moving body 122 will come into contact, and when there is no possibility that the first moving body 120 and the second moving body 122 will come into contact with each other. The risk level is set to 0. In this embodiment, when the traveling trajectory (planned) of the first moving body 120 and the traveling trajectory (estimated) of the second moving body 122 intersect, the first moving body 120 and the second moving body 122 come into contact. It is determined that there is a possibility. Furthermore, in the present embodiment, when the traveling direction (planned) of the first mobile body 120 after passing the intersection 132 and the traveling direction (estimated) of the second mobile body 122 after passing the intersection 132 match, , it is determined that there is a possibility that the first moving body 120 and the second moving body 122 will come into contact.

The mobile body control unit 76 performs travel control related to automatic control according to the action plan. For example, the mobile body control unit 76 calculates the acceleration for causing the first mobile body 120 to travel at the speed V determined by the action planning unit 74. Furthermore, the mobile object control section 76 calculates a steering angle for causing the first mobile object 120 to travel along the traveling trajectory determined by the action planning section 74. The mobile body control unit 76 outputs instruction information related to automatic control (driving instructions, braking instructions, steering instructions) to the drive device 28, the braking device 30, and the steering device 32. The notification control unit 78 outputs instruction information (notification instruction) to the HMI 34 when notification occurs in the action plan.

[1.1.5. Scene-Danger Information 88]
The scene-risk information 88 stored in the storage device 60 will be explained using FIG. 3 and FIGS. 4A to 4E. The scene-risk information 88 is related to a scene defined by the shape of the road 130, the position of the first moving body 120, and the position of the second moving body 122, and the contact between the first moving body 120 and the second moving body 122. Correspond with the degree of risk.

The scene-risk information 88 shown in FIG. 3 includes scene information 90 and risk information 92. The scene information 90 includes first information 94 to fifth information 102. The first information 94 is information indicating the form of the road 130, specifically the intersection 132, as shown in FIG. 4A and the like. The second information 96 is information indicating the position of the first moving body 120, specifically, the position of the lane in which the first moving body 120 travels before entering the intersection 132. The third information 98 is information indicating the traveling direction of the first moving body 120, specifically, the direction in which the first moving body 120 passes through the intersection 132 and advances. The fourth information 100 is information indicating the position of the second moving body 122, specifically, the position of the lane in which the second moving body 122 travels before entering the intersection 132. The fifth information 102 is information indicating the traveling direction of the second moving body 122, specifically, the lighting direction of the turn indicator of the second moving body 122. The degree of risk information 92 is information indicating the degree of risk assumed from the scene indicated by the scene information 90.

[1.2. Overview of this embodiment]
An overview of this embodiment will be explained using FIG. 5. When the first moving body 120 passes through the intersection 132, the estimation unit 86 estimates the degree of risk for each second moving body 122. At this time, the estimation unit 86 changes the way of estimating the degree of risk according to the separation distance D between the first moving body 120 and the second moving body 122.

When the separation distance D between the first moving body 120 and the second moving body 122 is less than or equal to the first distance D1, the estimating unit 86 determines whether the second moving body 122 is dangerous based on its behavior (velocity, acceleration, and direction of movement). Estimate degree. This estimation is called the first estimation.

When the separation distance D between the first moving body 120 and the second moving body 122 is greater than or equal to the second distance D2 (>D1), the estimation unit 86 estimates the degree of risk based on the scene at that time. This estimation is called second estimation.

When the separation distance D between the first moving body 120 and the second moving body 122 is greater than the first distance D1 and less than the second distance D2, the estimation unit 86 performs the first estimation and the second estimation, and Select the one with higher confidence among the two estimation results. This estimation is called the third estimation.

[1.3. Processing performed by estimation device 36]
[1.3.1. Main processing]
The main processing will be explained using FIG. 6. The main processing shown in FIG. 6 is executed at predetermined time intervals.

In step S1, the estimation unit 86 performs an execution determination process to determine whether it is time to execute the estimation process (step S4). The execution determination process will be explained in [1.3.2] (FIG. 7) below. When the process of step S1 is completed, the process moves to step S2.

In step S2, the distance acquisition unit 84 obtains the speed V of the first moving object 120 based on the detection result of the speed sensor (vehicle behavior sensor 22). Then, the distance acquisition unit 84 acquires a first distance D1 and a second distance D2, which is longer than the first distance D1, based on the speed V. The first distance D1 is the sum of the first predetermined distance and the first variable distance. The second distance D2 is the sum of the second predetermined distance and the second variable distance. The first predetermined distance and the second predetermined distance are stored in the storage device 60 in advance. The first variable distance and the second variable distance are determined according to the speed V. For example, the distance acquisition unit 84 obtains the first variable distance and the second variable distance by multiplying the speed V by a predetermined coefficient. The predetermined coefficient used when determining the first variable distance and the predetermined coefficient used when determining the second variable distance may be the same or different. When the process of step S2 is completed, the process moves to step S3.

In step S3, the estimation unit 86 performs a D1 correction process to correct the first distance D1 calculated in step S2. The D1 correction process will be explained in [1.3.3] (FIG. 8) below. When the process in step S3 ends, the process moves to step S4.

In step S4, the estimation unit 86 performs estimation processing to estimate the degree of risk for each second moving object 122. The estimation process will be explained in [1.3.4] (FIG. 9) below. When the process of step S4 ends, the main process ends.

[1.3.2. Execution judgment process]
The execution determination process performed in step S1 of FIG. 6 will be explained using FIG. 7.

In step S<b>11 , the estimation unit 86 determines whether the first mobile object 120 is about to enter the intersection 132 based on the recognition result of the self-position recognition unit 72 . If the first mobile object 120 is about to enter the intersection 132 (step S11: YES), the process moves to step S12. On the other hand, if the first moving body 120 is not in a situation where it enters the intersection 132 (step S11: NO), the main process shown in FIG. 6 ends.

In step S12, the estimation unit 86 determines whether or not there is a second moving object 122 traveling toward the intersection 132, based on the recognition result of the external world recognition unit 70. If the second moving body 122 is present (step S12: YES), the process moves to step S13. On the other hand, if the second moving body 122 is not present (step S12: NO), the main process shown in FIG. 6 ends.

In step S13, the estimation unit 86 determines, based on the recognition result of the self-position recognition unit 72, whether the position of the first moving body 120 is a position at which risk estimation is to be performed (estimated execution position). In this embodiment, the estimated execution position is the boundary position Pb of the intersection 132. The boundary position Pb here is the position of the boundary through which the first moving body 120 passes when entering the intersection 132 from outside to inside. If the position of the first moving body 120 is the estimated execution position (step S13: YES), the execution determination process ends and the process moves to step S2 in FIG. 6. On the other hand, if the position of the first moving body 120 is not the estimated execution position (step S13: NO), the main process shown in FIG. 6 ends.

[1.3.3. D1 correction processing]
The D1 correction process performed in step S3 of FIG. 6 will be explained using FIG. 8.

In step S21, the estimating unit 86 counts the number of second moving bodies 122 (number N of moving bodies) traveling in the oncoming lane 134 within the first distance D1 based on the recognition result of the external world recognizing unit 70. When the process of step S21 ends, the process moves to step S22.

In step S22, the estimation unit 86 compares the number N of moving objects with a predetermined threshold Nth. If the number N of moving objects is equal to or greater than the threshold value Nth (step S22: YES), the process moves to step S23. On the other hand, if the number N of moving objects is less than the threshold value Nth (step S22: NO), the D1 correction process ends and the process moves to step S4 in FIG. 6. In this case, the first distance D1 is maintained without being corrected.

In step S23, the estimation unit 86 corrects the first distance D1. If there are many second moving objects 122 within the first distance D1, the estimation unit 86 needs to perform the first estimation for many second moving objects 122. This increases the calculation load on the calculation device 58. Therefore, in order to reduce the calculation load, the estimation unit 86 reduces the number of second moving objects 122 to which the first estimation is performed by reducing the first distance D1. A predetermined reduction rate is stored in the storage device 60, and the estimation unit 86 reduces the first distance D1 by multiplying the first distance D1 by the reduction rate. The reduction rate may be a constant value or may be a value determined according to the number N of moving objects. For example, as the number N of moving objects increases, the reduction rate also increases. When the process in step S23 ends, the D1 correction process ends, and the process moves to step S4 in FIG. 6.

[1.3.4. Estimation processing]
The estimation process performed in step S4 of FIG. 6 will be explained using FIG. 9.

In step S31, the estimating unit 86 recognizes the separation distance D between the first moving body 120 and all the recognized second moving bodies 122 based on the recognition result of the external world recognizing unit 70. When the process of step S31 ends, the process moves to step S32.

In step S32, the estimation unit 86 determines which of the first estimation to the third estimation should be used to estimate the degree of risk for each second moving object 122. If the separation distance D is less than or equal to the first distance D1 (step S32: D≦D1), the process moves to step S33. If the separation distance D is greater than or equal to the second distance D2 (step S32: D≧D2), the process moves to step S34. If the separation distance D is greater than the first distance D1 and less than the second distance D2 (step S32: D1<D<D2), the process moves to step S35.

In step S33, the estimation unit 86 determines the degree of risk based on the highly accurate first estimation. Here, an example of the first estimation will be explained. The estimating unit 86 estimates the traveling trajectory of the second moving body 122 when passing through the intersection 132 based on the optical flow of the second moving body 122 and the operation status of the direction indicator. Furthermore, the estimating unit 86 generates a travel trajectory (plan) when the first mobile object 120 passes through the intersection 132. When the two travel trajectories intersect or are close to each other, the estimation unit 86 determines the time for the second mobile body 122 to reach the intersection position or the proximity position of both travel trajectories based on the speed and acceleration of the second mobile body 122. . Similarly, the estimating unit 86 calculates the time required for the first moving body 120 to reach the intersection position or the proximity position of both traveling trajectories based on the speed V and acceleration of the first moving body 120. Then, the estimating unit 86 determines the time difference between the two times, and estimates the risk level as "1" if the time difference is within a predetermined time, and estimates the risk level as "0" if the time difference is greater than or equal to the predetermined time. presume. Note that the estimating unit 86 estimates the risk level to be "0" even when the two travel trajectories do not intersect and are not close to each other. When the process of step S33 ends, the estimation process ends.

In step S34, the estimating unit 86 determines the degree of risk using the second estimation, which requires less calculation load. Here, an example of the second estimation will be explained. The estimation unit 86 determines the scene at that time in the following manner. The estimation unit 86 specifies the form of the road 130 (intersection 132) based on the recognition result of the external world recognition unit 70 or the information in the map DB 50. Furthermore, the estimating unit 86 identifies the lane in which the first mobile object 120 travels based on the recognition result of the external world recognition unit 70 or the recognition result of the self-position recognition unit 72. Furthermore, the estimating unit 86 specifies the traveling direction of the first moving body 120 according to the current driving situation (for example, whether the vehicle is traveling toward a destination or traveling along a road). Further, the estimating unit 86 identifies the lane in which the second mobile object 122 travels and the operating status of the turn signal indicator based on the recognition result of the external world recognizing unit 70 . The estimating unit 86 identifies the scene as described above, and compares the identified scene with the scene information 90 of the scene-risk information 88. Then, the estimating unit 86 takes the degree of risk associated with the scene information 90 that matches the specified scene as the estimation result. When the process of step S34 ends, the estimation process ends.

In step S35, the estimation unit 86 determines the degree of risk based on the third estimation. As described above, the third estimation is an estimation method in which the first estimation and the second estimation are performed, and when the two estimation results are different, the one with higher reliability is selected. The reliability may be determined by artificial intelligence or based on a predetermined pattern. For example, the estimating unit 86 basically determines that the reliability of the first estimation is high, and determines that the reliability of the second estimation is high when it corresponds to an exceptional pattern. An example of an exceptional pattern is when the degree of deviation from the existing scene, for example, the amount of deviation of the center of the second moving body 122 from the center of the lane is larger than a predetermined value, or the time change of the amount of deviation is greater than a predetermined value. There are cases where it is larger than. Further, as an example of an exceptional pattern, there is a case where the degree of recognition of the target second moving object 122 is low (such as a case where there are few recognizable parts). The exceptional pattern is stored in the storage device 60 in advance. When the process of step S35 ends, the estimation process ends.

[2. Modified example]
There are various modifications to the embodiments described above. Some of the modified examples will be explained below. Note that the embodiment described above and each modification described below can be combined or replaced as appropriate.

[2.1. Modification example 1: Execution determination process]
In step S1 shown in FIG. 6, instead of performing the execution determination process shown in FIG. 7, the execution determination process shown in FIG. 10 may be performed. Among the processes shown in FIG. 10, the processes in step S42 and step S43 are the same as the processes in step S12 and step S13 shown in FIG. Therefore, only step S41 will be explained below.

In step S41, the estimation unit 86 determines whether the first moving body 120 is in a situation where it crosses the extension range 136 of the oncoming lane 134. That is, the estimating unit 86 determines whether the first mobile object 120 turns right at the intersection 132 when driving on the left, and determines whether the first mobile object 120 turns left at the intersection 132 when driving on the right. Determine whether If the first moving body 120 is in a situation where it crosses the extension range 136 of the oncoming lane 134 (step S41: YES), the process moves to step S42. On the other hand, if the first moving body 120 is not in a situation where it crosses the extension range 136 of the oncoming lane 134 (step S41: NO), the main process shown in FIG. 6 ends.

[2.2. Modification 2: Target lane determination processing]
In step S4 shown in FIG. 6, the target lane determination process shown in FIG. 11 may be performed instead of the estimation process shown in FIG. 9 being performed.

In step S51, the estimation unit 86 determines whether the first moving object 120 can enter the target lane 138. Specifically, the estimation unit 86 determines whether or not there is a third mobile object 124 that stops within a predetermined range of the target lane 138. The target lane 138 is a lane into which the first moving object 120 enters after leaving the intersection 132. The predetermined range is a range within a predetermined distance from the entrance of the target lane 138, that is, the boundary between the intersection 132 and the target lane 138 in the direction of travel. As the predetermined distance, for example, a length equal to or less than the vehicle length of the first moving body 120 is set. If the third moving body 124 stops within the predetermined range and the first moving body 120 cannot enter the target lane 138 (step S51: NO), the process moves to step S52. A specific example is a situation where the target lane 138 is congested. On the other hand, if the third moving body 124 does not stop within the predetermined range and the first moving body 120 can enter the target lane 138 (step S51: YES), the process moves to step S53.

In step S52, the estimating unit 86 determines the degree of risk for all second moving objects 122 by second estimation. The first moving object 120 stops when it cannot enter the target lane 138. In this case, the degree of risk estimation accuracy does not need to be high. For this reason, the estimation unit 86 does not perform the first estimation with high accuracy, but preferentially performs the second estimation with a small calculation load.

In step S53, the estimation unit 86 performs the estimation process shown in FIG. 9.

[2.3. Modification 3]
In the embodiment described above, the estimation execution position at which the estimation unit 86 estimates the degree of risk is the boundary position Pb of the intersection 132. However, the estimated execution position can be set as appropriate. For example, the estimating unit 86 performs the first estimation of the degree of risk when the first moving object 120 approaches the intersection 132 and reaches the boundary position Pb, and the first moving object 120 A second estimate of risk may be made when crossing the road. When the estimation unit 86 performs estimation multiple times when passing through the intersection 132, the estimation unit 86 may use the estimation result of the second estimation in the first estimation in the second and subsequent estimations.

[2.4. Modification example 4]
When the oncoming lane 134 is congested, the estimation unit 86 may estimate the degree of risk only for the second moving body 122 whose separation distance D is less than or equal to the first distance D1. In this case, the estimation unit 86 determines the degree of risk based on the first estimation. The estimation unit 86 determines whether there is a traffic jam based on the speed or density of the second moving body 122.

[2.5. Modification 5]
In the embodiment described above, the first mobile body 120 including the estimation device 36 has a driving support function or an automatic driving function. However, the first mobile body 120 including the estimation device 36 does not need to have a driving support function or an automatic driving function. In this case, the degree of risk is used for purposes other than determinations related to travel of the first moving body 120. For example, the degree of danger is notified to the occupants.

[3. Second embodiment]
In the embodiment and modification described above, the estimating unit 86 changes the way of estimating the degree of risk depending on the separation distance D between the first moving body 120 and the second moving body 122. Alternatively, the estimating unit 86 may change the method of estimating the degree of risk depending on the expected approach time between the first moving body 120 and the second moving body 122.

In this case, as shown in FIG. 12, the action planning section 74 includes a time calculation section 184 and an estimation section 86 that perform processing related to estimating the degree of risk. The time calculating unit 184 calculates, for each second moving object 122, the expected approach time when the first moving object 120 and the oncoming lane 134 approach.

The estimation unit 86 estimates the degree of risk of the second mobile object 122 whose expected approach time is less than or equal to the first time using the first estimation. The estimating unit 86 estimates the degree of risk using the second estimation for the second moving object 122 whose estimated approach time is a second time or more, which is longer than the first time. Furthermore, the estimating unit 86 estimates the degree of risk using the third estimation for the second moving body 122 whose estimated approach time is longer than the first time and less than the second time.

Also in the second embodiment, modifications of the first embodiment can be applied as appropriate.

[4. Other embodiments]
In the first embodiment and its modified examples, the estimation unit 86 does not need to perform the third estimation. For example, when the separation distance D between the first moving body 120 and the second moving body 122 is greater than the first distance D1 and less than the second distance D2, the estimation unit 86 may perform the first estimation. , a second estimation may be performed. Similarly, in the second embodiment and its modified examples, the estimation unit 86 does not need to perform the third estimation. For example, if the estimated time of approach between the first moving body 120 and the second moving body 122 is longer than the first time and less than the second time, the estimating unit 86 may perform the first estimation; A second estimation may also be performed.

[5. Technical ideas obtained from the embodiment]
The technical ideas that can be understood from the above embodiments and modified examples will be described below.

The first aspect of the present invention is
an external world recognition unit 70 that recognizes the environment around the moving object (first moving object 120);
a distance acquisition unit 84 that acquires a first distance D1 from the first moving body 120 and a second distance D2 that is longer than the first distance;
A scene defined by the shape of the road 130, the position of the mobile body, and the position of a traffic participant (second mobile body 122) around the mobile body, and the degree of risk regarding contact between the mobile body and the traffic participant. a scene storage unit (storage device 60) that stores and in association with each other;
For the traffic participant whose separation distance D between the mobile object and the traffic participant is less than or equal to the first distance D1, the speed of the traffic participant, the acceleration of the traffic participant, and the traffic participant's A first estimation is performed to estimate the degree of risk based on the traveling direction, and for the traffic participants whose separation distance D is greater than or equal to the second distance D2, the degree of risk is estimated based on the scene at that time. an estimation unit 86 that performs a second estimation to estimate the degree;
Equipped with

In the above configuration, the estimating unit 86 performs a highly accurate first estimation for the second moving body 122 whose separation distance D from the first moving body 120 is short; For the second moving body 122 whose distance D is a long distance, a second estimation with a small calculation load is performed. In this way, the estimation unit 86 estimates the second moving object 122 that requires highly accurate estimation (the second moving object 122 whose separation distance D is a short distance) and the second moving object that does not require highly accurate estimation. 122 (second moving body 122 whose separation distance D is a long distance), the method of estimating the degree of risk is changed. Therefore, according to the above configuration, it is possible to reduce the load of calculations performed for estimating the degree of risk regarding contact between a moving object (first moving object 120) and a traffic participant (second moving object 122).

In the first aspect of the present invention,
The estimating unit 86 calculates the first estimation and the above for the traffic participant (second moving object 122) whose separation distance D is larger than the first distance D1 and less than the second distance D2. A second estimation may be performed, and the estimation result of the first estimation and the estimation result of the second estimation may be selected as the estimation result of the degree of risk.

In the first aspect of the present invention,
The traffic participant (second moving body 122) moves toward the intersection 132 in the oncoming lane 134,
The estimating unit 86 performs the first estimation of the degree of risk when the mobile object (first mobile object 120) reaches the boundary position Pb of the intersection 132 from outside the intersection 132, and The degree of risk may be estimated a second time when crossing the extended range 136 of the oncoming lane 134, and the estimation result of the second estimation obtained in the first estimation may be used in the second estimation.

According to the above configuration, since the estimation result of the second estimation in the first estimation is also used in the second estimation, the computational load of the second estimation is reduced.

In the first aspect of the present invention,
The traffic participant (second moving body 122) moves toward the intersection 132 in the oncoming lane 134,
The estimation unit 86 calculates the estimation unit 86 in a situation where the moving object (first moving object 120) crosses the extension range 136 of the oncoming lane 134 at the intersection 132 and enters the target lane 138 beyond the intersection 132. After estimating the degree of risk, if the mobile object cannot enter the target lane 138 because there is another traffic participant (third mobile object 124) stopping within a predetermined range of the target lane 138, The second estimation may be performed instead of the first estimation also for the traffic participant (second moving body 122) for which the separation distance D is less than or equal to the first distance D1.

According to the above configuration, the estimating unit 86 performs the second estimation with a small calculation load on the nearby second moving object 122 in a situation where highly accurate estimation is not required. Therefore, according to the above configuration, the calculation load of the risk estimation process can be reduced.

In the first aspect of the present invention,
The estimating unit 86 determines whether the number of traffic participants (second moving objects 122) whose separation distance D is equal to or less than the first distance D1 (number of moving objects N) is greater than a predetermined number (threshold value Nth). , the degree of risk may be estimated after changing the first distance D1 to a short distance.

If there are many second moving objects 122 that are the targets of the first estimation, the accuracy of the first estimation will decrease. By changing the first distance D1 to a shorter distance as in the above configuration, the number of second moving objects 122 that are the targets of the first estimation can be reduced. As a result, it is possible to suppress a decrease in the accuracy of the first estimation.

In the first aspect of the present invention,
The first distance D1 is the sum of a first predetermined distance and a first variable distance determined according to the speed V of the moving body (first moving body 120),
The second distance D2 may be a sum of a second predetermined distance longer than the first predetermined distance and a second variable distance determined according to the speed V of the moving object.

The second aspect of the present invention is
an external world recognition unit 70 that recognizes the environment around the moving object (first moving object 120);
a time calculation unit 184 that calculates an expected approach time when the mobile object and a traffic participant (second mobile object 122) on the opposite lane 134 side approach;
a scene storage unit that stores a scene defined by the shape of the road 130, the position of the moving body, and the position of the traffic participant, and a degree of risk regarding contact between the moving body and the traffic participant, in association with each other; storage device 60);
For the traffic participant whose scheduled approach time is less than or equal to a first time, the risk level is estimated based on the speed of the traffic participant, the acceleration of the traffic participant, and the traveling direction of the traffic participant. A first estimation is performed to estimate the degree of risk based on the scene at that time for the traffic participant whose estimated approach time is a second time or more, which is longer than the first time. An estimation unit 86 that performs estimation;
Equipped with

According to the above configuration, the calculation load of the risk estimation process can be reduced.

In a second aspect of the invention,
The estimating unit 86 uses the first estimation and the second estimation for the traffic participant (second moving object 122) whose estimated approach time is longer than the first time and less than the second time. Estimation may be performed, and one of the estimation results of the first estimation and the estimation result of the second estimation may be selected as the estimation result of the risk level.

Note that the estimation device according to the present invention is not limited to the above-described embodiments, and can of course take various configurations without departing from the gist of the present invention.

36... Estimation device 60... Storage device (scene storage unit)
70...External world recognition section 84...Distance acquisition section 86...Estimation section 120...First moving object (moving object)
122...Second mobile body (traffic participant) 124...Third mobile body (other traffic participant)
130...Road 132...Intersection 134...Oncoming lane 136...Extension range 138...Target lane 184...Time calculation unit

Claims (8)

an external world recognition unit that recognizes the surrounding environment of the moving object;
a distance acquisition unit that acquires a first distance from the moving body and a second distance longer than the first distance;
A scene defined by a road configuration, a position of the moving object, a position of a traffic participant around the moving object, and a degree of risk regarding contact between the moving object and the traffic participant are stored in association with each other. a scene memory section;
For the traffic participant whose separation distance between the mobile object and the traffic participant is less than or equal to the first distance, the speed of the traffic participant, the acceleration of the traffic participant, and the traveling direction of the traffic participant. A first estimation is performed to estimate the degree of risk based on the second distance, and for the traffic participants whose separation distance is greater than or equal to the second distance, the degree of risk is estimated based on the scene at that time. A second estimation is performed, and for the traffic participants whose separation distance is greater than the first distance and less than the second distance, at least one of the first estimation and the second estimation is performed. Department and
An estimation device comprising: The estimation device according to claim 1,
The estimating unit performs the first estimation and the second estimation for the traffic participants whose separation distance is greater than the first distance and less than the second distance, and An estimating device that selects the one with higher reliability between the estimation result and the second estimation result as the estimation result of the risk level. The estimation device according to claim 2,
the traffic participant moves towards the intersection in the oncoming lane;
The estimating unit performs the first estimation of the degree of risk when the moving object reaches a boundary position of the intersection from outside the intersection, and when the moving object crosses an extension range of the oncoming lane. An estimation device that performs a second estimation of the degree of risk and uses the estimation result of the second estimation in the first estimation also in the second estimation. The estimation device according to claim 1,
the traffic participant moves towards the intersection in the oncoming lane;
The estimating unit estimates the degree of risk in a situation where the moving object crosses the extension range of the oncoming lane at the intersection and enters the target lane beyond the intersection, If the mobile object cannot enter the target lane because there is another traffic participant stopping within the same distance, the first distance is also applied to the traffic participant whose separation distance is less than or equal to the first distance. An estimation device that performs the second estimation instead of estimation. The estimation device according to any one of claims 1 to 4,
The estimating unit changes the first distance to a shorter distance and estimates the degree of risk when the number of traffic participants whose separation distance is equal to or less than the first distance is greater than a predetermined number. Estimation device. The estimation device according to any one of claims 1 to 5,
The first distance is a sum of a first predetermined distance and a first variable distance determined according to the speed of the moving object,
The estimating device, wherein the second distance is a sum of a second predetermined distance longer than the first predetermined distance and a second variable distance determined according to the speed of the moving object. an external world recognition unit that recognizes the surrounding environment of the moving object;
a time calculation unit that calculates a scheduled approach time for the moving object and a traffic participant on the opposite lane to approach;
A scene defined by a road configuration, a position of the moving object, a position of the traffic participant around the moving object, and a degree of risk regarding contact between the moving object and the traffic participant are stored in association with each other. a scene memory section to
For the traffic participant whose scheduled approach time is less than or equal to a first time, the risk level is estimated based on the speed of the traffic participant, the acceleration of the traffic participant, and the traveling direction of the traffic participant. A first estimation is performed to estimate the degree of risk based on the scene at that time for the traffic participant whose estimated approach time is a second time or more, which is longer than the first time. estimation, and for the traffic participants whose estimated approach time is longer than the first time and less than the second time, at least one of the first estimation and the second estimation is performed. Department and
An estimation device comprising: The estimation device according to claim 7,
The estimating unit performs the first estimation and the second estimation for the traffic participant whose estimated approach time is longer than the first time and less than the second time, and is configured to perform the first estimation and the second estimation. An estimating device that selects the one with higher reliability between the estimation result of the second estimation and the estimation result of the second estimation, and sets it as the estimation result of the risk level.

Images